Architect, Machine Learning
Ramesh Sridharan is a Machine Learning Architect at Manifold, where he collaborates with customers to understand their business challenges and design production-grade solutions. Ramesh focuses on picking the right tool for the job, from a random forest to deep neural networks. Recent product successes include designing an optimization algorithm for a medical device company, and predicting network outages across networks and applications for a networking hardware company.
Prior to Manifold, Ramesh led a machine learning engineering team at Vidado. There, he led a team of engineers that developed the industry’s most accurate handwriting recognition, using tools such as PyTorch, Keras, Tensorflow, and AWS. In addition to training models and conducting computer vision research, he developed the company’s data acquisition and labeling strategy and implemented processes for accelerating machine research to production. Ramesh also serves as a Lecturer at UC Berkeley, where his course “Foundations of Data Science” teaches hundreds of undergraduates the fundamentals of programming, data visualization, statistics, and machine learning.
Ramesh has a PhD in Computer Science from MIT, specializing in machine learning applied to medical imaging; he has a BS in Computer Science from UC Berkeley.